A genome-wide atlas of human cell morphology.

A key challenge of the modern genomics era is developing data-driven representations of gene function. Here, we present the first unbiased morphology-based genome-wide perturbation atlas in human cells, containing three genome-scale genotype-phenotype maps comprising >20,000 single-gene CRISPR-Cas9-based knockout experiments in >30 million cells. Our optical pooled cell profiling approach (PERISCOPE) combines a de-stainable high-dimensional phenotyping panel (based on Cell Painting1,2) with optical sequencing of molecular barcodes and a scalable open-source analysis pipeline to facilitate massively parallel screening of pooled perturbation libraries. This approach provides high-dimensional phenotypic profiles of individual cells, while simultaneously enabling interrogation of subcellular processes. Our atlas reconstructs known pathways and protein-protein interaction networks, identifies culture media-specific responses to gene knockout, and clusters thousands of human genes by phenotypic similarity. Using this atlas, we identify the poorly-characterized disease-associated transmembrane protein TMEM251/LYSET as a Golgi-resident protein essential for mannose-6-phosphate-dependent trafficking of lysosomal enzymes, showing the power of these representations. In sum, our atlas and screening technology represent a rich and accessible resource for connecting genes to cellular functions at scale.

The microtubule regulator ringer functions downstream from the RNA repair/splicing pathway to promote axon regeneration.

Promoting axon regeneration in the central and peripheral nervous system is of clinical importance in neural injury and neurodegenerative diseases. Both pro- and antiregeneration factors are being identified. We previously reported that the Rtca mediated RNA repair/splicing pathway restricts axon regeneration by inhibiting the nonconventional splicing of Xbp1 mRNA under cellular stress. However, the downstream effectors remain unknown. Here, through transcriptome profiling, we show that the tubulin polymerization-promoting protein (TPPP) ringmaker/ringer is dramatically increased in Rtca-deficient Drosophila sensory neurons, which is dependent on Xbp1. Ringer is expressed in sensory neurons before and after injury, and is cell-autonomously required for axon regeneration. While loss of ringer abolishes the regeneration enhancement in Rtca mutants, its overexpression is sufficient to promote regeneration both in the peripheral and central nervous system. Ringer maintains microtubule stability/dynamics with the microtubule-associated protein futsch/MAP1B, which is also required for axon regeneration. Furthermore, ringer lies downstream from and is negatively regulated by the microtubule-associated deacetylase HDAC6, which functions as a regeneration inhibitor. Taken together, our findings suggest that ringer acts as a hub for microtubule regulators that relays cellular status information, such as cellular stress, to the integrity of microtubules in order to instruct neuroregeneration.

Systematic identification of cancer cell vulnerabilities to natural killer cell-mediated immune surveillance.

Only a subset of cancer patients respond to T-cell checkpoint inhibitors, highlighting the need for alternative immunotherapeutics. We performed CRISPR-Cas9 screens in a leukemia cell line to identify perturbations that enhance natural killer effector functions. Our screens defined critical components of the tumor-immune synapse and highlighted the importance of cancer cell interferon-γ signaling in modulating NK activity. Surprisingly, disrupting the ubiquitin ligase substrate adaptor DCAF15 strongly sensitized cancer cells to NK-mediated clearance. DCAF15 disruption induced an inflamed state in leukemic cells, including increased expression of lymphocyte costimulatory molecules. Proteomic and biochemical analysis revealed that cohesin complex members were endogenous client substrates of DCAF15. Genetic disruption of DCAF15 was phenocopied by treatment with indisulam, an anticancer drug that functions through DCAF15 engagement. In AML patients, reduced DCAF15 expression was associated with improved survival. These findings suggest that DCAF15 inhibition may have useful immunomodulatory properties in the treatment of myeloid neoplasms.

The ER membrane protein complex interacts cotranslationally to enable biogenesis of multipass membrane proteins.

The endoplasmic reticulum (ER) supports biosynthesis of proteins with diverse transmembrane domain (TMD) lengths and hydrophobicity. Features in transmembrane domains such as charged residues in ion channels are often functionally important, but could pose a challenge during cotranslational membrane insertion and folding. Our systematic proteomic approaches in both yeast and human cells revealed that the ER membrane protein complex (EMC) binds to and promotes the biogenesis of a range of multipass transmembrane proteins, with a particular enrichment for transporters. Proximity-specific ribosome profiling demonstrates that the EMC engages clients cotranslationally and immediately following clusters of TMDs enriched for charged residues. The EMC can remain associated after completion of translation, which both protects clients from premature degradation and allows recruitment of substrate-specific and general chaperones. Thus, the EMC broadly enables the biogenesis of multipass transmembrane proteins containing destabilizing features, thereby mitigating the trade-off between function and stability.

Lipid Homeostasis Is Maintained by Dual Targeting of the Mitochondrial PE Biosynthesis Enzyme to the ER.

Spatial organization of phospholipid synthesis in eukaryotes is critical for cellular homeostasis. The synthesis of phosphatidylcholine (PC), the most abundant cellular phospholipid, occurs redundantly via the ER-localized Kennedy pathway and a pathway that traverses the ER and mitochondria via membrane contact sites. The basis of the ER-mitochondrial PC synthesis pathway is the exclusive mitochondrial localization of a key pathway enzyme, phosphatidylserine decarboxylase Psd1, which generates phosphatidylethanolamine (PE). We find that Psd1 is localized to both mitochondria and the ER. Our data indicate that Psd1-dependent PE made at mitochondria and the ER has separable cellular functions. In addition, the relative organellar localization of Psd1 is dynamically modulated based on metabolic needs. These data reveal a critical role for ER-localized Psd1 in cellular phospholipid homeostasis, question the significance of an ER-mitochondrial PC synthesis pathway to cellular phospholipid homeostasis, and establish the importance of fine spatial regulation of lipid biosynthesis for cellular functions.

The SND proteins constitute an alternative targeting route to the endoplasmic reticulum.

In eukaryotes, up to one-third of cellular proteins are targeted to the endoplasmic reticulum, where they undergo folding, processing, sorting and trafficking to subsequent endomembrane compartments. Targeting to the endoplasmic reticulum has been shown to occur co-translationally by the signal recognition particle (SRP) pathway or post-translationally by the mammalian transmembrane recognition complex of 40 kDa (TRC40) and homologous yeast guided entry of tail-anchored proteins (GET) pathways. Despite the range of proteins that can be catered for by these two pathways, many proteins are still known to be independent of both SRP and GET, so there seems to be a critical need for an additional dedicated pathway for endoplasmic reticulum relay. We set out to uncover additional targeting proteins using unbiased high-content screening approaches. To this end, we performed a systematic visual screen using the yeast Saccharomyces cerevisiae, and uncovered three uncharacterized proteins whose loss affected targeting. We suggest that these proteins work together and demonstrate that they function in parallel with SRP and GET to target a broad range of substrates to the endoplasmic reticulum. The three proteins, which we name Snd1, Snd2 and Snd3 (for SRP-independent targeting), can synthetically compensate for the loss of both the SRP and GET pathways, and act as a backup targeting system. This explains why it has previously been difficult to demonstrate complete loss of targeting for some substrates. Our discovery thus puts in place an essential piece of the endoplasmic reticulum targeting puzzle, highlighting how the targeting apparatus of the eukaryotic cell is robust, interlinked and flexible.

Finding FISH in a small pond.

Advanced microscopy and labeling techniques reveal that bacteria localize mRNAs within their cells in a similar way to eukaryotes.

LOCAL TRANSLATION. Response to Comment on "Principles of ER cotranslational translocation revealed by proximity-specific ribosome profiling".

Reid and Nicchitta propose that most cellular translation is carried out by a noncycling pool of endoplasmic reticulum (ER)-associated ribosomes. However, proximity-specific ribosome profiling data place an upper bound of about 7 to 16% on the fraction of cytosolic protein translation carried out by ribosomes accessible to ER-tethered biotin ligases. Moreover, yeast pulse-labeling experiments argue against there being a static population of ER-associated ribosomes.

Targeting and plasticity of mitochondrial proteins revealed by proximity-specific ribosome profiling.

Nearly all mitochondrial proteins are nuclear-encoded and are targeted to their mitochondrial destination from the cytosol. Here, we used proximity-specific ribosome profiling to comprehensively measure translation at the mitochondrial surface in yeast. Most inner-membrane proteins were cotranslationally targeted to mitochondria, reminiscent of proteins entering the endoplasmic reticulum (ER). Comparison between mitochondrial and ER localization demonstrated that the vast majority of proteins were targeted to a specific organelle. A prominent exception was the fumarate reductase Osm1, known to reside in mitochondria. We identified a conserved ER isoform of Osm1, which contributes to the oxidative protein-folding capacity of the organelle. This dual localization was enabled by alternative translation initiation sites encoding distinct targeting signals. These findings highlight the exquisite in vivo specificity of organellar targeting mechanisms.

Principles of ER cotranslational translocation revealed by proximity-specific ribosome profiling.

Localized protein synthesis is a fundamental mechanism for creating distinct subcellular environments. Here we developed a generalizable proximity-specific ribosome profiling strategy that enables global analysis of translation in defined subcellular locations. We applied this approach to the endoplasmic reticulum (ER) in yeast and mammals. We observed the large majority of secretory proteins to be cotranslationally translocated, including substrates capable of posttranslational insertion in vitro. Distinct translocon complexes engaged nascent chains at different points during synthesis. Whereas most proteins engaged the ER immediately after or even before signal sequence (SS) emergence, a class of Sec66-dependent proteins entered with a looped SS conformation. Finally, we observed rapid ribosome exchange into the cytosol after translation termination. These data provide insights into how distinct translocation mechanisms act in concert to promote efficient cotranslational recruitment.

Global analyses of the effect of different cellular contexts on microRNA targeting.

MicroRNA (miRNA) regulation clearly impacts animal development, but the extent to which development-with its resulting diversity of cellular contexts-impacts miRNA regulation is unclear. Here, we compared cohorts of genes repressed by the same miRNAs in different cell lines and tissues and found that target repertoires were largely unaffected, with secondary effects explaining most of the differential responses detected. Outliers resulting from differential direct targeting were often attributable to alternative 3' UTR isoform usage that modulated the presence of miRNA sites. More inclusive examination of alternative 3' UTR isoforms revealed that they influence ∼10% of predicted targets when comparing any two cell types. Indeed, considering alternative 3' UTR isoform usage improved prediction of targeting efficacy significantly beyond the improvements observed when considering constitutive isoform usage. Thus, although miRNA targeting is remarkably consistent in different cell types, considering the 3' UTR landscape helps predict targeting efficacy and explain differential regulation that is observed.

Extensive alternative polyadenylation during zebrafish development.

The post-transcriptional fate of messenger RNAs (mRNAs) is largely dictated by their 3' untranslated regions (3' UTRs), which are defined by cleavage and polyadenylation (CPA) of pre-mRNAs. We used poly(A)-position profiling by sequencing (3P-seq) to map poly(A) sites at eight developmental stages and tissues in the zebrafish. Analysis of over 60 million 3P-seq reads substantially increased and improved existing 3' UTR annotations, resulting in confidently identified 3' UTRs for >79% of the annotated protein-coding genes in zebrafish. mRNAs from most zebrafish genes undergo alternative CPA, with those from more than a thousand genes using different dominant 3' UTRs at different stages. These included one of the poly(A) polymerase genes, for which alternative CPA reinforces its repression in the ovary. 3' UTRs tend to be shortest in the ovaries and longest in the brain. Isoforms with some of the shortest 3' UTRs are highly expressed in the ovary, yet absent in the maternally contributed RNAs of the embryo, perhaps because their 3' UTRs are too short to accommodate a uridine-rich motif required for stability of the maternal mRNA. At 2 h post-fertilization, thousands of unique poly(A) sites appear at locations lacking a typical polyadenylation signal, which suggests a wave of widespread cytoplasmic polyadenylation of mRNA degradation intermediates. Our insights into the identities, formation, and evolution of zebrafish 3' UTRs provide a resource for studying gene regulation during vertebrate development.

Conserved function of lincRNAs in vertebrate embryonic development despite rapid sequence evolution.

Thousands of long intervening noncoding RNAs (lincRNAs) have been identified in mammals. To better understand the evolution and functions of these enigmatic RNAs, we used chromatin marks, poly(A)-site mapping and RNA-Seq data to identify more than 550 distinct lincRNAs in zebrafish. Although these shared many characteristics with mammalian lincRNAs, only 29 had detectable sequence similarity with putative mammalian orthologs, typically restricted to a single short region of high conservation. Other lincRNAs had conserved genomic locations without detectable sequence conservation. Antisense reagents targeting conserved regions of two zebrafish lincRNAs caused developmental defects. Reagents targeting splice sites caused the same defects and were rescued by adding either the mature lincRNA or its human or mouse ortholog. Our study provides a roadmap for identification and analysis of lincRNAs in model organisms and shows that lincRNAs play crucial biological roles during embryonic development with functionality conserved despite limited sequence conservation.

Formation, regulation and evolution of Caenorhabditis elegans 3'UTRs.

Post-transcriptional gene regulation frequently occurs through elements in mRNA 3' untranslated regions (UTRs). Although crucial roles for 3'UTR-mediated gene regulation have been found in Caenorhabditis elegans, most C. elegans genes have lacked annotated 3'UTRs. Here we describe a high-throughput method for reliable identification of polyadenylated RNA termini, and we apply this method, called poly(A)-position profiling by sequencing (3P-Seq), to determine C. elegans 3'UTRs. Compared to standard methods also recently applied to C. elegans UTRs, 3P-Seq identified 8,580 additional UTRs while excluding thousands of shorter UTR isoforms that do not seem to be authentic. Analysis of this expanded and corrected data set suggested that the high A/U content of C. elegans 3'UTRs facilitated genome compaction, because the elements specifying cleavage and polyadenylation, which are A/U rich, can more readily emerge in A/U-rich regions. Indeed, 30% of the protein-coding genes have mRNAs with alternative, partially overlapping end regions that generate another 10,480 cleavage and polyadenylation sites that had gone largely unnoticed and represent potential evolutionary intermediates of progressive UTR shortening. Moreover, a third of the convergently transcribed genes use palindromic arrangements of bidirectional elements to specify UTRs with convergent overlap, which also contributes to genome compaction by eliminating regions between genes. Although nematode 3'UTRs have median length only one-sixth that of mammalian 3'UTRs, they have twice the density of conserved microRNA sites, in part because additional types of seed-complementary sites are preferentially conserved. These findings reveal the influence of cleavage and polyadenylation on the evolution of genome architecture and provide resources for studying post-transcriptional gene regulation.

A single Hox locus in Drosophila produces functional microRNAs from opposite DNA strands.

MicroRNAs (miRNAs) are approximately 22-nucleotide RNAs that are processed from characteristic precursor hairpins and pair to sites in messages of protein-coding genes to direct post-transcriptional repression. Here, we report that the miRNA iab-4 locus in the Drosophila Hox cluster is transcribed convergently from both DNA strands, giving rise to two distinct functional miRNAs. Both sense and antisense miRNA products target neighboring Hox genes via highly conserved sites, leading to homeotic transformations when ectopically expressed. We also report sense/antisense miRNAs in mouse and find antisense transcripts close to many miRNAs in both flies and mammals, suggesting that additional sense/antisense pairs exist.

Intronic microRNA precursors that bypass Drosha processing.

MicroRNAs (miRNAs) are approximately 22-nucleotide endogenous RNAs that often repress the expression of complementary messenger RNAs. In animals, miRNAs derive from characteristic hairpins in primary transcripts through two sequential RNase III-mediated cleavages; Drosha cleaves near the base of the stem to liberate a approximately 60-nucleotide pre-miRNA hairpin, then Dicer cleaves near the loop to generate a miRNA:miRNA* duplex. From that duplex, the mature miRNA is incorporated into the silencing complex. Here we identify an alternative pathway for miRNA biogenesis, in which certain debranched introns mimic the structural features of pre-miRNAs to enter the miRNA-processing pathway without Drosha-mediated cleavage. We call these pre-miRNAs/introns 'mirtrons', and have identified 14 mirtrons in Drosophila melanogaster and another four in Caenorhabditis elegans (including the reclassification of mir-62). Some of these have been selectively maintained during evolution with patterns of sequence conservation suggesting important regulatory functions in the animal. The abundance of introns comparable in size to pre-miRNAs appears to have created a context favourable for the emergence of mirtrons in flies and nematodes. This suggests that other lineages with many similarly sized introns probably also have mirtrons, and that the mirtron pathway could have provided an early avenue for the emergence of miRNAs before the advent of Drosha.

Large-scale sequencing reveals 21U-RNAs and additional microRNAs and endogenous siRNAs in C. elegans.

We sequenced approximately 400,000 small RNAs from Caenorhabditis elegans. Another 18 microRNA (miRNA) genes were identified, thereby extending to 112 our tally of confidently identified miRNA genes in C. elegans. Also observed were thousands of endogenous siRNAs generated by RNA-directed RNA polymerases acting preferentially on transcripts associated with spermatogenesis and transposons. In addition, a third class of nematode small RNAs, called 21U-RNAs, was discovered. 21U-RNAs are precisely 21 nucleotides long, begin with a uridine 5'-monophosphate but are diverse in their remaining 20 nucleotides, and appear modified at their 3'-terminal ribose. 21U-RNAs originate from more than 5700 genomic loci dispersed in two broad regions of chromosome IV-primarily between protein-coding genes or within their introns. These loci share a large upstream motif that enables accurate prediction of additional 21U-RNAs. The motif is conserved in other nematodes, presumably because of its importance for producing these diverse, autonomously expressed, small RNAs (dasRNAs).

A two-hit trigger for siRNA biogenesis in plants.

In Arabidopsis, microRNA-directed cleavage can define one end of RNAs that then generate phased siRNAs. However, most miRNA-targeted RNAs do not spawn siRNAs, suggesting the existence of additional determinants within those that do. We find that in moss, phased siRNAs arise from regions flanked by dual miR390 cleavage sites. AtTAS3, an siRNA locus important for development and conserved among higher plants, also has dual miR390 complementary sites. Both sites bind miR390 in vitro and are functionally required in Arabidopsis, but cleavage is undetectable at the 5' site--demonstrating that noncleavable sites can be functional in plants. Phased siRNAs also emanate from the bounded regions of every Arabidopsis gene with two known microRNA/siRNA complementary sites, but only rarely from genes with single sites. Therefore, two "hits,"--often, but not always, two cleavage events--constitute a conserved trigger for siRNA biogenesis, a finding with implications for recognition and silencing of aberrant RNA.

A consensus CaMK IV-responsive RNA sequence mediates regulation of alternative exons in neurons.

Neurons make extensive use of alternative pre-mRNA splicing to regulate gene expression and diversify physiological responses. We showed previously in a pituitary cell line that the Ca(++)/calmodulin-dependent protein kinase CaMK IV specifically repressed splicing of the BK channel STREX exon. This repression is dependent on a CaMK IV-responsive RNA element (CaRRE) within the STREX 3' splice site. Here, we report that similar Ca(++) regulation of splicing, mediated by L-type calcium channels and CaM kinase IV, occurs in cultured neurons and in the brain. We identify a critical CaRRE motif (CACATNRTTAT) that is essential for conferring CaMK IV repression on an otherwise constitutive exon. Additional Ca(++)-regulated exons that carry this consensus sequence are also identified in the human genome. Thus, the Ca(++)/CaMK IV pathway in neurons controls the alternative splicing of a group of exons through this short CaRRE consensus sequence. The functions of some of these exons imply that splicing control through the CaMK IV pathway will alter neuronal activity.

The widespread impact of mammalian MicroRNAs on mRNA repression and evolution.

Thousands of mammalian messenger RNAs are under selective pressure to maintain 7-nucleotide sites matching microRNAs (miRNAs). We found that these conserved targets are often highly expressed at developmental stages before miRNA expression and that their levels tend to fall as the miRNA that targets them begins to accumulate. Nonconserved sites, which outnumber the conserved sites 10 to 1, also mediate repression. As a consequence, genes preferentially expressed at the same time and place as a miRNA have evolved to selectively avoid sites matching the miRNA. This phenomenon of selective avoidance extends to thousands of genes and enables spatial and temporal specificities of miRNAs to be revealed by finding tissues and developmental stages in which messages with corresponding sites are expressed at lower levels.